Over a decade ago, we demonstrated that the latent viral reservoir in the resting CD4+ T cell compartment persists in virtually all HIV-infected individuals receiving clinically effective ART. In addition, we demonstrated that HIV continually replicates at low levels in chronically infected individuals who are consistently aviremic during prolonged periods of receiving ART. Consequently, the persistent viral reservoir has become a major impediment to the eradication of HIV in infected individuals receiving ART. In recent years, major research efforts have been dedicated to a better understanding of the pathogenesis of persistent HIV infection and to the development of therapeutic strategies aimed at eradicating virus in infected individuals receiving ART. During the past year, we have focused our research on: 1) examining the effect of ART on HIV reservoirs in elite controllers and 2) developing immune-based therapeutic strategies that target persistently infected CD4+ T cells. First, we investigated the effect of ART on the persistent HIV reservoirs in elite controllers. Although ART suppresses plasma viremia to below the limit of detection in the vast majority of HIV-infected individuals, the virus cannot be eradicated by therapy alone due in part to the persistence of viral reservoirs in the peripheral blood, lymphoid tissues, and other sequestered sites. Consequently, there is a growing interest in developing therapeutic strategies to eliminate persistent HIV reservoirs and/or to enhance host immunity against the virus in order to control viral replication in the absence of ART. In this regard, it has been shown that a small proportion of HIV-infected individuals spontaneously control plasma viremia in the absence of ART (controllers). Moreover, a subset of such individuals (elite controllers) are capable of maintaining relatively normal CD4+ T cell counts and undetectable levels of plasma viremia for years to decades without the need for ART. Although previous studies have suggested that ongoing viral replication may occur in elite controllers, precise frequencies of infected CD4+ T cells carrying replication-competent virus and/or the impact of ART on immunologic and virologic parameters have not been fully delineated in this patient population. Therefore, we conducted a study in order to determine the effect of ART on the dynamics of persistent HIV reservoirs and various immunologic parameters in elite controllers. We demonstrated that a short course (9 months) of ART dramatically decreased the infectious HIV burden in elite controllers, providing compelling evidence that onging/residual viral replication occurs in such aviremic individuals. The infectious HIV burden in the CD4+ T cell compartment of the elite controllers rebounded back to their original baseline levels upon cessation of ART, an indication that a virologic set-point exists even at this very low level of viral replication, and that host immunity can efficiently control plasma viremia without completely eliminating the infected CD4+ T cells. We also demonstrated that the level of HIV-specific CD8+ T cells in elite controllers gradually declined upon initiation of ART, further supporting the concept that low levels of viral replication occur in these infected individuals. Our data provide compelling evidence that ongoing HIV replication occurs in elite controllers. Given that these individuals effectively control viral replication and lack disease progression in the absence of ART, our study has implications for the feasibility of attaining sustained virologic remission in non-controllers without complete suppression or elimination of viral replication in the face of discontinuation of ART. Second, we investigated the feasibility of eliminating HIV-infected CD4+ T cells using autologous CD8+ T cells that express a chimeric antigen receptor (CAR). Recent studies have demonstrated that complete remission can be achieved in patients with chronic lymphoid leukemia upon infusion of autologous CD8+ T cells expressing CAR with specificity for the B cell antigen CD19 coupled with CD3-zeta, and costimulatory molecules. Inspired by these studies, we investigated whether autologous CD8+ T cells expressing CAR specific for HIV Env could eliminate persistently infected CD4+ T cells from HIV-infected individuals receiving ART. Three CAR constructs were evaluated. All constructs included CD4 for recognition of HIV Env on infected targets, CD3-zeta for cytotoxic activity, and one of the following constimulatory molecules: mock, 4-1BB, or CD28. Autologous CD8+ T cells expressing the above CAR proteins were cocultured with pre-stimulated CD4+ T cells from HIV-infected individuals receiving ART and the level of viral replication was monitored by p24 ELISA over time. Our preliminary data suggest that the highest level of HIV suppression can be achieved when CD8+ T cells expressing the CAR that contains CD28 are cocultured with infected target cells. Our data also show that CD8+ T cells expressing the above CAR construct produce higher levels of intracellular cytokines and chemokines (such as IFN-g, TNF-a, MIP-1) compared to those expressing CAR with mock or 4-1BB upon incubation with a cell line expressing HIV Env. Taken together, our preliminary data suggest that the elimination of persistently infected CD4+ T cells from HIV-infected individuals ex vivo can be achieved with autologous CD8+ T cells that express a CAR that recognizes HIV Env. Because CD8+ cytotoxic T cells do not require viral antigen specificity (besides recognition of HIV Env) for killing, such effector cells could potentially eliminate infected target cells without inducing viral escape mutants. In addition, considering the longevity of CAR-transduced CD8+ T cells observed in cancer patients (years to decades), these cells could restrict viral replication in vivo for extended periods of time.